PROF. C. G. ABBOT has contributed to the Scientific Monthly (November, 1917) a reasoned discussion, in the light of recent investigations, of the extent and probable sequence of the effect of solar variation on world weather. More than one independent line of argument will be found to point to the conclusion that in a period of two thousand years there has been no appreciable change of climate. Therefore the balance of the heat exchanges between the earth's income from the solar radiation and its expenditure in terrestrial radiation into space may be regarded as only fluctuating between narrow limits. Eighty per cent, of the solar radiation fails to reach the earth's surface through its protecting envelopes, and 90 per cent, of the terrestrial radiation fails to escape. Such is the beneficent effect of our atmosphere, for want of which the temperature of the moon's surface, as proved by actual observation, falls during the short period of a lunar eclipse many times as far as does that of any part of the earth between day and night. In most places on the earth the surface air temperature rarely varies as much as 1 per cent, from day to day, but the variation between day and night is affected by the character of the surface, Timbuktu, in the Sahara desert, having twice the daily and four times the annual change of temperature at Port au Prince, Haiti, in approximately the same latitude.